Multi-Agent Reinforcement Learning (MARL) is a subfield of reinforcement learning that is becoming increasingly relevant and has been blowing my mind. Multi-agent reinforcement learning studies how multiple agents interact in a common environment. That is, when these agents interact with the environment and one another, can we observe them collaborate, coordinate, compete, or collectively learn to accomplish a particular task.
We are a group of Reinforcement learning Enthusiasts, dwelling into the realms of Multi Agent Reinforcement Learning. We have weekly / bi weekly sessions of Paper Reading where our main two goals are:
Each meeting has a main topic of discussion. a topic which rotates among the participants’ choices. The discussion consists of multiple people sharing their insights and a summarization of related papers. The main goal is to familiarize ourselves with the maximum number of related concepts in the same subdomain. Each presentation is followed by doubt sessions and thinking sessions while notes are maintained for future reference.
All participants are supposed to read the selected paper carefully, writing down comments and questions they would like to discuss in the meeting. This helps us look for opportunities to improve in the paper, and pursue them as a research project.
The Reading group is open to motivated students. If you are interested in participating, contact Somnath Sendhil Kumar.
We are very thankful to the OpenSource Community Specially, Lantao Yu and Co. for providing a wonderful starting ground for searching papers in Multi Agent Reinforcement Learning. You can check it out Here MARL-Papers
The members of the COPS IG group are actively working in the multiple fields and some of the domains of expertise are :
So we would be happy to collaborate, and have interesting talks with you.